HyperAIHyperAI超神经

Command Palette

Search for a command to run...

算力平台

Stock Market Prediction On Astock

评估指标

Accuray
F1-score
Precision
Recall

评测结果

各个模型在此基准测试上的表现结果

Paper Title
SRL&SDPG&Factors75.4075.1275.4275.23FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model
SDPG&Factors73.1272.9773.0472.96FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model
SRL&Factors69.4869.2869.5469.41FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model
Chinese Pert Large (News+Factors)67.3767.2767.2867.73PERT: Pre-training BERT with Permuted Language Model
SRL66.8966.9266.9266.95Astock: A New Dataset and Automated Stock Trading based on Stock-specific News Analyzing Model
Chinese Lert Large (News+Factors)66.3666.1666.4066.69LERT: A Linguistically-motivated Pre-trained Language Model
SRL&SDPG66.1066.0166.0466.09FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model
Chinese Pert Large (News)65.0965.0365.0265.07PERT: Pre-training BERT with Permuted Language Model
Chinese Lert Large (News)64.3764.3064.3464.31LERT: A Linguistically-motivated Pre-trained Language Model
Factos Only63.7463.6663.6763.71FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model
RoBERTa WWM Ext (News+Factors)62.4962.5462.5962.51RoBERTa: A Robustly Optimized BERT Pretraining Approach
RoBERTa WWM Ext (News)61.3461.4861.9761.32RoBERTa: A Robustly Optimized BERT Pretraining Approach
XLNET Chinese61.1461.1961.6061.09Revisiting Pre-Trained Models for Chinese Natural Language Processing
ERNIE-SKEP60.6660.6661.8560.59SKEP: Sentiment Knowledge Enhanced Pre-training for Sentiment Analysis
Bert Chinese59.1158.9959.0759.20BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
HAN Stock57.3556.6158.4157.20Listening to Chaotic Whispers: A Deep Learning Framework for News-oriented Stock Trend Prediction
StockNet46.7244.4447.6546.68Stock Movement Prediction from Tweets and Historical Prices
0 of 17 row(s) selected.